A new survey from Zapier puts a number on something many business leaders have been quietly feeling for months: the AI skills crisis is real, it is urgent, and most companies are doing almost nothing structured to fix it.

The survey, conducted in late March 2026 by research firm Centiment across 542 C-level executives and decision-makers at US organisations with more than 1,000 employees, found that 77% consider workforce AI skills an urgent need. That number would be reassuring if it were matched by action. It is not.

The Gap Between Urgency and Action

Nearly all of the executives surveyed — 94% — confirmed their organisations are already using AI in some capacity. So the tools are in the building. The capability to use them well is not.

Formal AI training reaches only:

• 51% of IT and engineering teams
• 43% of sales and marketing staff
• 40% of HR professionals
• Fewer still in finance, legal, and operations

That means roughly half of even the most AI-adjacent teams are going into this era without any structured support. The rest of the business is largely on its own.

Only 7% of the executives surveyed said their Learning and Development or HR teams are responsible for AI training initiatives. Most organisations are leaving it to individual employees to figure out, or to the IT department to handle as a technical rollout rather than a capability-building exercise.

What Happens When Training Doesn’t Happen

The consequences of unstructured AI adoption are showing up in the data. Workers who have not received training are six times more likely to say AI makes them less productive, not more.

That is a striking finding. It suggests the tool gap is almost irrelevant. Companies can buy every AI licence available, deploy every agent, and roll out every copilot, and still see declining output if their people do not know how to work alongside these systems effectively.

The research also found that 78% of executives report their organisations face at least one significant barrier to building AI skills. The most common barrier, cited by 18%, is that the pace of AI change makes training feel obsolete before it can be deployed. That is a real challenge, but it is also an argument for continuous, embedded learning rather than one-off training events.

What This Means for Business

The picture here is clear: AI capability is becoming a competitive divide, and most companies are on the wrong side of it. A WRITER survey of 2,400 executives and workers found that 60% of companies plan layoffs for employees who refuse to adopt AI — so the urgency is not just economic, it is existential for individual roles. The data on how data-literate companies outperform their peers shows this gap has compounded for years — AI is just accelerating it.

The organisations that get this right are not the ones that deploy the most tools. They are the ones building systematic fluency at every level of the organisation. Gartner has separately noted that companies using AI as a form of “people amplification” outperform those using it purely for headcount reduction. Structured upskilling is the mechanism that makes amplification possible.

The challenge for many businesses is that traditional training approaches were built for slower-moving skills. You cannot run a quarterly workshop on AI fundamentals when the fundamentals shift monthly. What works is learning embedded in daily workflows, structured so that people build capability as they do actual work, not in a classroom removed from it. (A realistic look at what that progression from Excel to AI actually looks like makes this concrete.)

This is precisely the gap that platforms like Enterprise DNA Learn are designed to close. The issue is not awareness — 77% of executives already understand the urgency. The issue is building the infrastructure to act on it, at scale, in a way that keeps pace with how fast the technology is changing.

For any business owner or people leader reading this: if you cannot answer the question “who specifically owns AI upskilling in our organisation, and what does their plan look like?” — the survey results suggest you are in the majority. And that majority is falling behind.

If your team is ready to close that gap with structured, practical data and AI training, Enterprise DNA Learn gives your people the skills that translate into real outcomes — not another certification they complete once and forget.